[gentoo-user] Whats better for crossplatform applications?

[Alexander Pilipovsky]

May be, it's not a "only Gentoo" question, but I want to write and start
applications under Gentoo and Windows. I saw Tcl/Tk library in work (as
example OOMMF: http://math.nist.gov/oommf/, but it, sometimes, unstable
under Windows XP). And it did not like me to look of buttons, lists etc.
Other way I saw in using wxPython (http://www.wxpython.org/) or
wxWidgets (http://www.wxwidgets.org/). I want to have as little as
possible differences in GUI of my program when it starts under GNOME,
KDE or Windows. May be some other libraries for crossplatform
development are exists.

What library better for unification of application look and developing?

P. S. Excuse my English, please :)

-- 
Alexander Pilipovsky aka Engraver




signature.asc
Description: OpenPGP digital signature

Re: [gentoo-user] grub: how to install new version of stage 1

[Alan McKinnon]

On Saturday 30 May 2009 20:59:00 John P. Burkett wrote:
> The manual suggests doing "grub-install --no-floppy /dev/sda"
> but later says "If your system does not have any floppy drives, add the
> --no-floppy option to the above command to prevent grub from probing the
> (non-existing) floppy drives."  My machine has a floppy drive. Should I
> omit the --no-floppy option and just do "grub-install /dev/sda" ?

The manual is actually quite clear if you know even just a little bit about 
boot loaders.

Use --no-floppy if

a) you do not have a floppy drive
b) you do not intend grub to use the floppy drive you do have

The question you should be asking is "have I ever booted off a floppy drive in 
the last X years, and do I ever intend do so again?"

The first example in the manual is assuming the answers are no and no - pretty 
normal for the vast majority of users.

-- 
alan dot mckinnon at gmail dot com

Re: [gentoo-user] Re: Disabling swap & mounting /tmp on tmpfs = new standard?

[Volker Armin Hemmann]

On Sonntag 31 Mai 2009, Grant wrote:
>  I recently disabled swap and mounted /tmp on tmpfs for a netbook since
>  the SSD is so slow, and now I'm wondering if that would be a wise move
>  for all of my Gentoo systems.  In what type of situation would it be a
>  bad idea?
> >>>
> >>> Instead of disabling swap, just make it small (like 32MB or something;
> >>> whatever the smallest allowable partition size is).  The kernel needs
> >>> swap to operate optimally, even if it's extremely small.  Just make
> >>> sure it's there.
> >>
> >> Hmm, on this old box I noticed swap was using more than 135,000K earlier
> >> today as I was emerging xulrunner and ImageMagick.  I think that the
> >> size of swap is relevant to the memory size that the box in question
> >> has.  Not all machines have found their way to 2G RAM yet ... ;)
> >
> > Don't forget that you can set swapiness too.  This is set in
> > /etc/sysctl.conf and for mine I have this:
> >
> > vm.swappiness = 30
> >
> > The lower the number, the less chance of it using swap.  If it is set to
> > 90, it will use a lot of swap which is fine if you have little ram or a
> > really fast drive.  If it is set to 30, then it will not use swap unless
> > it is basically out of ram.
> >
> > With the setting of 30, mine uses swap when compiling OOo or some other
> > large package or if I am opening a TON of pics.  Otherwise, swap is at 0
> > or close to it even after being up a long time.  I have 2Gbs here tho.
> > Your mileage may vary.
> >
> > Dale
>
> Thanks Dale.  Should "vm.swappiness = 30" work well on all systems?
> My Gentoo systems have vastly different specs and duties so I love
> tweaks that always improve things.  It sounds like /tmp on tmpfs is
> one of those.
>
> - Grant

you can set swappiness = 0 which works even better, because the kernel will 
only swap if it really has too.

Re: [gentoo-user] get fences failed: -1 and [drm:i915_getparam] *ERROR*

[AJ Spagnoletti]

> http://en.gentoo-wiki.com/wiki/Intel_GMA
>
> which is pertinent to my hardware and kernel/drivers. Therefore if I do
>
>
> -> vblank_mode=0 glxgears
> ATTENTION: default value of option vblank_mode overridden by environment.
> get fences failed: -1
> param: 6, val: 0
> 4418 frames in 5.0 seconds = 883.510 FPS
> 4490 frames in 5.0 seconds = 897.871 FPS
> 4491 frames in 5.0 seconds = 898.054 FPS
> 4481 frames in 5.0 seconds = 896.043 FPS
> 4382 frames in 5.0 seconds = 876.251 FPS
>
> I get much better performance as I used to have before the xorg upgrade.
> This is where I am now trying to gather  information on how to deal with
> dri and vblank_mode settings.
>
> Hope this helps.
>
> --
> Valmor
>
>

Using the link you posted I was able to get similar numbers on my
card, if you scroll down to the bottom of the page there where it
talks about driconf, and set up your drirc file you should be able to
apply these settings system wide. I am still have some issues with
compiz-fusion running smoothly I currently dont have the cube working
it lags very bad with that. I will do some more research into the
kernel bug and see if upgrading and re-enabling tiling fixes this
issue. Thanks for the link and pointing me in the right direction.

AJ

Re: [gentoo-user] Re: Disabling swap & mounting /tmp on tmpfs = new standard?

[Dale]

Grant wrote:
> I recently disabled swap and mounted /tmp on tmpfs for a netbook since
> the SSD is so slow, and now I'm wondering if that would be a wise move
> for all of my Gentoo systems.  In what type of situation would it be a
> bad idea?
>
>   
 Instead of disabling swap, just make it small (like 32MB or something;
 whatever the smallest allowable partition size is).  The kernel needs
 swap to operate optimally, even if it's extremely small.  Just make sure
 it's there.

 
>>> Hmm, on this old box I noticed swap was using more than 135,000K earlier 
>>> today
>>> as I was emerging xulrunner and ImageMagick.  I think that the size of swap
>>> is relevant to the memory size that the box in question has.  Not all
>>> machines have found their way to 2G RAM yet ... ;)
>>>
>>>
>>>   
>> Don't forget that you can set swapiness too.  This is set in
>> /etc/sysctl.conf and for mine I have this:
>>
>> vm.swappiness = 30
>>
>> The lower the number, the less chance of it using swap.  If it is set to
>> 90, it will use a lot of swap which is fine if you have little ram or a
>> really fast drive.  If it is set to 30, then it will not use swap unless
>> it is basically out of ram.
>>
>> With the setting of 30, mine uses swap when compiling OOo or some other
>> large package or if I am opening a TON of pics.  Otherwise, swap is at 0
>> or close to it even after being up a long time.  I have 2Gbs here tho.
>> Your mileage may vary.
>>
>> Dale
>> 
>
> Thanks Dale.  Should "vm.swappiness = 30" work well on all systems?
> My Gentoo systems have vastly different specs and duties so I love
> tweaks that always improve things.  It sounds like /tmp on tmpfs is
> one of those.
>
> - Grant
>
>
>   

I'm on x86 and I really don't know where the vm part came from.  It
could have been me that put it there but I think it may have gotten
updated somewhere along the way.  I'm not sure what would update that
tho.  May be worth a google for your arch and just swappiness and see
what else can be in front of it. 

I used to have it set to 70 when I only had 512MBs of ram.  It would use
swap pretty regular, even just for caching stuff.  So, the setting does
work for sure.  If you wanted it to use swap only to prevent the system
from crashing, I would assume you could set it to 10 or something like
that.  If you have a really fast drive, SATA or something, then you
could set it to 90 and let it use swap all it wants.

Dale

:-)  :-) 

Re: [gentoo-user] Re: Disabling swap & mounting /tmp on tmpfs = new standard?

[Grant]

 I recently disabled swap and mounted /tmp on tmpfs for a netbook since
 the SSD is so slow, and now I'm wondering if that would be a wise move
 for all of my Gentoo systems.  In what type of situation would it be a
 bad idea?

>>> Instead of disabling swap, just make it small (like 32MB or something;
>>> whatever the smallest allowable partition size is).  The kernel needs
>>> swap to operate optimally, even if it's extremely small.  Just make sure
>>> it's there.
>>>
>>
>> Hmm, on this old box I noticed swap was using more than 135,000K earlier 
>> today
>> as I was emerging xulrunner and ImageMagick.  I think that the size of swap
>> is relevant to the memory size that the box in question has.  Not all
>> machines have found their way to 2G RAM yet ... ;)
>>
>>
>
> Don't forget that you can set swapiness too.  This is set in
> /etc/sysctl.conf and for mine I have this:
>
> vm.swappiness = 30
>
> The lower the number, the less chance of it using swap.  If it is set to
> 90, it will use a lot of swap which is fine if you have little ram or a
> really fast drive.  If it is set to 30, then it will not use swap unless
> it is basically out of ram.
>
> With the setting of 30, mine uses swap when compiling OOo or some other
> large package or if I am opening a TON of pics.  Otherwise, swap is at 0
> or close to it even after being up a long time.  I have 2Gbs here tho.
> Your mileage may vary.
>
> Dale

Thanks Dale.  Should "vm.swappiness = 30" work well on all systems?
My Gentoo systems have vastly different specs and duties so I love
tweaks that always improve things.  It sounds like /tmp on tmpfs is
one of those.

- Grant

Re: [gentoo-user] Re: Disabling swap & mounting /tmp on tmpfs = new standard?

[Dale]

Mick wrote:
> On Saturday 30 May 2009, Nikos Chantziaras wrote:
>   
>> Grant wrote:
>> 
>>> I recently disabled swap and mounted /tmp on tmpfs for a netbook since
>>> the SSD is so slow, and now I'm wondering if that would be a wise move
>>> for all of my Gentoo systems.  In what type of situation would it be a
>>> bad idea?
>>>   
>> Instead of disabling swap, just make it small (like 32MB or something;
>> whatever the smallest allowable partition size is).  The kernel needs
>> swap to operate optimally, even if it's extremely small.  Just make sure
>> it's there.
>> 
>
> Hmm, on this old box I noticed swap was using more than 135,000K earlier 
> today 
> as I was emerging xulrunner and ImageMagick.  I think that the size of swap 
> is relevant to the memory size that the box in question has.  Not all 
> machines have found their way to 2G RAM yet ... ;)
>
>   

Don't forget that you can set swapiness too.  This is set in
/etc/sysctl.conf and for mine I have this:

vm.swappiness = 30

The lower the number, the less chance of it using swap.  If it is set to
90, it will use a lot of swap which is fine if you have little ram or a
really fast drive.  If it is set to 30, then it will not use swap unless
it is basically out of ram.

With the setting of 30, mine uses swap when compiling OOo or some other
large package or if I am opening a TON of pics.  Otherwise, swap is at 0
or close to it even after being up a long time.  I have 2Gbs here tho. 
Your mileage may vary.

Dale

:-)  :-) 

Re: [gentoo-user] [OT] Running two apaches and MySQLs on the same server

[Mick]

On Thursday 28 May 2009, Alan McKinnon wrote:
> On Thursday 28 May 2009 21:51:26 Stroller wrote:
> > > So I recommend option 4:
> > >
> > > Pony up the money for server #2
> >
> > Just for the sake of satanic advocacy, could you indulge me, please?
> >
> > Let's say Mick is the administrator for all domains in question. He
> > decides to run the two sites on different machines, one for
> > MickBlog.org and one for MicrophoneShoppe.com. If MickBlog is
> > insecure, what makes you think he will administer MicrophoneShoppe any
> > more securely?
>
> I suffer from a healthy dose of paranoia :-)

Well, it is commonly said that the fact you are paranoid doesn't necessarily 
mean they are not out to get you!  

> Added to that, my employer is an ISP and not shy with budgets, so a
> purchase order for new hardware in a case like this will not raise any
> eyebrows. For me, it's a low level of risk high impact scenario and the $
> cost is low.
>
> In a budget-constrained environment, it would obviously work very
> differently

Well, I am in a very cost constrained environment I'm afraid.  Good advice 
given here - I am now thinking that a virtual server is the next stage.  Any 
idea how it would run on a single CPU machine - or must we bite the bullet 
and go for some multicore monster?

> And yes, I do indeed not trust php code at all. I've seen the audit results
> of too many php projects that were diligently hardened and what it took to
> get them from working state to an acceptably secure state.

I haven't your specific experiences of course, but have read about and seen a 
few horror stories of cracked phpBB implementations that I know I would not 
be able to sleep at night ... especially as one of the hosted websites is 
running some home brew of php+perl.

Still, at least formally it is weak passwords that are usually blamed for most 
compromised servers.
-- 
Regards,
Mick


signature.asc
Description: This is a digitally signed message part.

Re: [gentoo-user] Re: Disabling swap & mounting /tmp on tmpfs = new standard?

[Mick]

On Saturday 30 May 2009, Nikos Chantziaras wrote:
> Grant wrote:
> > I recently disabled swap and mounted /tmp on tmpfs for a netbook since
> > the SSD is so slow, and now I'm wondering if that would be a wise move
> > for all of my Gentoo systems.  In what type of situation would it be a
> > bad idea?
>
> Instead of disabling swap, just make it small (like 32MB or something;
> whatever the smallest allowable partition size is).  The kernel needs
> swap to operate optimally, even if it's extremely small.  Just make sure
> it's there.

Hmm, on this old box I noticed swap was using more than 135,000K earlier today 
as I was emerging xulrunner and ImageMagick.  I think that the size of swap 
is relevant to the memory size that the box in question has.  Not all 
machines have found their way to 2G RAM yet ... ;)

-- 
Regards,
Mick


signature.asc
Description: This is a digitally signed message part.

[gentoo-user] emul-linux-x86-qtlibs for Qt4?

[Nikos Chantziaras]

emul-linux-x86-qtlibs only provides Qt3 libraries.  I need to run a Qt4 
32-bit app under Gentoo AMD64.  Is there a package somewhere in some 
overlay that offers 32-bit Qt4 libs?

Re: [gentoo-user] Sync'ed my ~x86 system yesterday and now resolver stopped working

[Timur Aydin]

Graham Murray wrote:

Timur Aydin  writes:

  

Hi,

I have synced my ~x86 system yesterday and after it completed, the
resolver doesn't work for some programs anymore. For example, ping
 says "unknown host name". It doesn't even contact the dns
server, which is running on the same host. But dig  works
fine. Also, using the IP address directly, I can access the internet.

I am suspecting that the new glibc 2.10 is causing this. Anybody else
having this issue?



I had this issue a couple of weeks ago. I think it was the upgrade to
net-dns/openresolv-3.3.2 which was responsible. The solution was to
edit etc/resolvconf.conf and uncomment the line
name_servers=127.0.0.1

  
That's what I tried yesterday and it resolved the problem. So it seems 
the new resolver does not default to checking localhost as a dns server 
and needs to be explicitely told to do so...


--
Timur

Re: Tweaks for SSDs [Was: [gentoo-user] [ot] no more inodes]

[Neil Bothwick]

On Sat, 30 May 2009 22:25:29 +0300, Eray Aslan wrote:

> > Or I could try btrfs, which has an ssd mount option.  
> 
> Ugh.  Even on-disk format is not finalized yet.

That's OK, I'm not using it on my backup server :)


-- 
Neil Bothwick

CAUTION: Do not install prior to installation.


signature.asc
Description: PGP signature

Re: [gentoo-user] grub: how to install new version of stage 1

[Dale]

John P. Burkett wrote:
> On a x86 machine I did "emerge -D -uav world" and got a response that
> read in part as follows:
>  * Messages for package sys-boot/grub-0.97-r9:
>  *
>  * To avoid automounting and autoinstalling with /boot,
>  * just export the DONT_MOUNT_BOOT variable.
>  *
>  * *** IMPORTANT NOTE: you must run grub and install
>  * the new version's stage1 to your MBR.  Until you do,
>  * stage1 and stage2 will still be the old version, but
>  * later stages will be the new version, which could
>  * cause problems such as an unbootable system.
>  * This means you must use either grub-install or perform
>  * root/setup manually! For more help, see the handbook:
>  *
> http://www.gentoo.org/doc/en/handbook/handbook-x86.xml?part=1&chap=10#grub-install-auto
>  * To interactively install grub files to another device such as a USB
>  * stick, just run the following and specify the directory as prompted:
>  *emerge --config =grub-0.97-r9
>  * Alternately, you can export GRUB_ALT_INSTALLDIR=/path/to/use to tell
>  * grub where to install in a non-interactive way.
>
> After reading
> http://www.gentoo.org/doc/en/handbook/handbook-x86.xml?part=1&chap=10#grub-install-auto
> I did "grep -v rootfs /proc/mounts > /etc/mtab". That seems to have
> produced the needed /etc/mtab file.  Now I'm confused by the part of the
> manual with code listings 2.6 and
> 2.7 and the associated commentary.
> The manual suggests doing "grub-install --no-floppy /dev/sda"
> but later says "If your system does not have any floppy drives, add the
> --no-floppy option to the above command to prevent grub from probing the
> (non-existing) floppy drives."  My machine has a floppy drive. Should I
> omit the --no-floppy option and just do "grub-install /dev/sda" ?
>
> -John
>
>
>   

I got something similar once before.  All I did was run grub, then while
inside grub, run root and setup.  Basically like I did when i first
installed Linux.  It has worked so far so I guess that was all that it
needed was to update the files in /boot for when it first loads after
the BIOS.

Hope that helps.

Dale

:-)  :-)

Re: Tweaks for SSDs [Was: [gentoo-user] [ot] no more inodes]

[Eray Aslan]

On Sat, May 30, 2009 at 02:40:34PM +0100, Neil Bothwick wrote:
> On Sat, 30 May 2009 12:06:04 +0200, Florian Philipp wrote:
> 
> > Delaying commits with ext4 and/or laptop-mode will reduce the wear-down
> > of your SSD but it might as well freeze your system when the actual
> > commit takes place because these things tend to have a terribly low
> > write performance.
> 
> That may explain the pauses I get from time to time. Maybe shortening the
> commit period will help.

Couple of points regarding the pauses, SSDs, schedulers and ext3/ext4:

* try ext4 with its delayed allocation. It should help with pauses
* ext3 with data=writeback should help.  Some security implications with
data=writeback tho.  So be careful if it is not a single user machine.
* Deadline scheduler has more throughput than CFQ or anticipatory but it
is totally unusable under load
* A lot of patches to ext3 and ext4 for a/m pauses and SSDs.  Some made
it to kernel 2.6.30 I believe.
* Try CFQ and NOOP as schedulers for SSDs for now.  After the above
patches, CFQ should be the better choice.

Basically, a lot of changes to ext3/ext4 and schedulers at the moment.
I would wait for at least kernel 2.6.31 before trying alternatives and
making decisions.
 
> Or I could try btrfs, which has an ssd mount option.

Ugh.  Even on-disk format is not finalized yet.

-- 
Eray

[gentoo-user] grub: how to install new version of stage 1

[John P. Burkett]

On a x86 machine I did "emerge -D -uav world" and got a response that
read in part as follows:
 * Messages for package sys-boot/grub-0.97-r9:
 *
 * To avoid automounting and autoinstalling with /boot,
 * just export the DONT_MOUNT_BOOT variable.
 *
 * *** IMPORTANT NOTE: you must run grub and install
 * the new version's stage1 to your MBR.  Until you do,
 * stage1 and stage2 will still be the old version, but
 * later stages will be the new version, which could
 * cause problems such as an unbootable system.
 * This means you must use either grub-install or perform
 * root/setup manually! For more help, see the handbook:
 *
http://www.gentoo.org/doc/en/handbook/handbook-x86.xml?part=1&chap=10#grub-install-auto
 * To interactively install grub files to another device such as a USB
 * stick, just run the following and specify the directory as prompted:
 *emerge --config =grub-0.97-r9
 * Alternately, you can export GRUB_ALT_INSTALLDIR=/path/to/use to tell
 * grub where to install in a non-interactive way.

After reading
http://www.gentoo.org/doc/en/handbook/handbook-x86.xml?part=1&chap=10#grub-install-auto
I did "grep -v rootfs /proc/mounts > /etc/mtab". That seems to have
produced the needed /etc/mtab file.  Now I'm confused by the part of the
manual with code listings 2.6 and
2.7 and the associated commentary.
The manual suggests doing "grub-install --no-floppy /dev/sda"
but later says "If your system does not have any floppy drives, add the
--no-floppy option to the above command to prevent grub from probing the
(non-existing) floppy drives."  My machine has a floppy drive. Should I
omit the --no-floppy option and just do "grub-install /dev/sda" ?

-John


-- 
John P. Burkett
Department of Economics
University of Rhode Island
Kingston, RI 02881-0808
USA

phone (401) 874-9195

[gentoo-user] Re: Disabling swap & mounting /tmp on tmpfs = new standard?

[Nikos Chantziaras]

Grant wrote:

I recently disabled swap and mounted /tmp on tmpfs for a netbook since
the SSD is so slow, and now I'm wondering if that would be a wise move
for all of my Gentoo systems.  In what type of situation would it be a
bad idea?


Instead of disabling swap, just make it small (like 32MB or something; 
whatever the smallest allowable partition size is).  The kernel needs 
swap to operate optimally, even if it's extremely small.  Just make sure 
it's there.


Mounting /tmp as tmpfs improves speed, so no problems there.  You might 
want to mount /var/tmp/portage as tmpfs too, that will give nice speed 
gains during emerge (if you have the RAM for it; a 2GB /var/tmp/portage 
should be enough for almost anything except OpenOffice, you'll have to 
umount to emerge that one.)

[gentoo-user] Disabling swap & mounting /tmp on tmpfs = new standard?

[Grant]

I recently disabled swap and mounted /tmp on tmpfs for a netbook since
the SSD is so slow, and now I'm wondering if that would be a wise move
for all of my Gentoo systems.  In what type of situation would it be a
bad idea?

- Grant

Re: [gentoo-user] Optimizations for SSD netbook

[Grant]

>> >> 2. added elevator=noop as a boot parameter
>> >
>> > I remember that I've given this second advice. Since then I've read in
>> > the German computer journal c't [1] that CFQ has a detection for SSDs
>> > since 2.6.28 and now is the best choice for these devices.
>>
>> OK, do I need a boot parameter if I've set CFQ as the default IO
>> scheduler in the kernel config?
>
> No, that's what default switch is there for.
>
>
>> > Yup, the entry should read:
>> > tmp /tmp tmpfs default 0 0
>
> I'd also suggest to explicitly specify max size of tmpfs mount, since
> system locking because of wrong cp command is probably the last thing
> you want. Argument is "size=" (see man 8 mount).
>
>
>> Do you think mounting /tmp in RAM is worthwhile?  Mike doesn't seem to
>> think too highly of it.
>
> I guess accelerated fsync and reduced disk wear should be a nice plus
> for SSD device, provided the path in question is constanly used for
> writing which really might be the case with files, created in /tmp by
> various mktemp implementations (like python's) which I haven't really
> thought about, so I think I might be wrong about the issue here, sorry.

Thanks guys.  I think the /tmp trick made a good difference.  The last
thing I can think of is pruning the kernel way down.  I think it's
mostly default.

- Grant

Re: [gentoo-user] Optimizations for SSD netbook

[Neil Bothwick]

On Sat, 30 May 2009 07:08:55 -0700, Grant wrote:

> Do you think mounting /tmp in RAM is worthwhile?  Mike doesn't seem to
> think too highly of it.

I do, especially on an SSD, but with any device it reduces disk access,
which is a good thing.


-- 
Neil Bothwick

April Fools! You're really in a holodeck simulation!


signature.asc
Description: PGP signature

Re: [gentoo-user] Optimizations for SSD netbook

[Mike Kazantsev]

On Sat, 30 May 2009 07:08:55 -0700
Grant  wrote:

> >> 2. added elevator=noop as a boot parameter
> >
> > I remember that I've given this second advice. Since then I've read in
> > the German computer journal c't [1] that CFQ has a detection for SSDs
> > since 2.6.28 and now is the best choice for these devices.
> 
> OK, do I need a boot parameter if I've set CFQ as the default IO
> scheduler in the kernel config?

No, that's what default switch is there for.


> > Yup, the entry should read:
> > tmp /tmp tmpfs default 0 0

I'd also suggest to explicitly specify max size of tmpfs mount, since
system locking because of wrong cp command is probably the last thing
you want. Argument is "size=" (see man 8 mount).


> Do you think mounting /tmp in RAM is worthwhile?  Mike doesn't seem to
> think too highly of it.

I guess accelerated fsync and reduced disk wear should be a nice plus
for SSD device, provided the path in question is constanly used for
writing which really might be the case with files, created in /tmp by
various mktemp implementations (like python's) which I haven't really
thought about, so I think I might be wrong about the issue here, sorry.


-- 
Mike Kazantsev // fraggod.net


signature.asc
Description: PGP signature

Re: [gentoo-user] Optimizations for SSD netbook

[Grant]

>> My girlfriend is at her wit's end with her SSD netbook and is now
>> hogging my laptop.  Her netbook has 1GB RAM that could be upgraded to
>> 1.5GB, but I've read that it's a pain.  It already runs xfce4, and
>> I've just made these optimizations based on past discussions:
>>
>> 1. CFLAGS="-march=prescott -0s -pipe -fomit-frame-pointer -ssse3"
>> 2. added elevator=noop as a boot parameter
>
> I remember that I've given this second advice. Since then I've read in
> the German computer journal c't [1] that CFQ has a detection for SSDs
> since 2.6.28 and now is the best choice for these devices.

OK, do I need a boot parameter if I've set CFQ as the default IO
scheduler in the kernel config?

>> 3. disabled DRI to save 32MB RAM
>> 4. removed the swap partition from /etc/fstab
>>
>> Am I missing anything significant?  I've read that it's good to set up
>> /tmp in RAM.  How can I do that? In /etc/fstab I have:
>>
>> shm /dev/shm tmpfs nodev,nosuid,noexec
>>
>> Is that related?
>>
>
> Yup, the entry should read:
> tmp /tmp tmpfs default 0 0

Do you think mounting /tmp in RAM is worthwhile?  Mike doesn't seem to
think too highly of it.

> The first entry is just a name. You can name it in every way you want it.
>
> For further tweaks: Do you use Firefox? I've read that it uses fsync()
> when writing to its sqlite backend. This is a really good thing because
> it reduces the risk of loosing data but you might (and can) disable this
> to increase performance and reduce wear. I don't have a link at hand but
> it shouldn't be to hard to find.

There is some interesting info on disabling fsync here:

http://www.flamingspork.com/projects/libeatmydata/

Sounds kinda dangerous. :)

- Grant

Re: Tweaks for SSDs [Was: [gentoo-user] [ot] no more inodes]

[Neil Bothwick]

On Sat, 30 May 2009 12:06:04 +0200, Florian Philipp wrote:

> Delaying commits with ext4 and/or laptop-mode will reduce the wear-down
> of your SSD but it might as well freeze your system when the actual
> commit takes place because these things tend to have a terribly low
> write performance.

That may explain the pauses I get from time to time. Maybe shortening the
commit period will help.

Or I could try btrfs, which has an ssd mount option.


-- 
Neil Bothwick

Electricians DO IT until it Hz...


signature.asc
Description: PGP signature

Re: [gentoo-user] pdftk emerge error

[Marco]

On Sat, May 30, 2009 at 7:02 AM, Dan Johansson  wrote:
> On Saturday 30 May 2009, Marco wrote:
>> Hi all,
>>
>> I am trying to emerge pdftk on my amd64 machine. After re-emerging gcc
>> with gcj, compilation fails with:
>>
>> * ERROR: app-text/pdftk-1.12 failed.
>>  * Call stack:
>>  *               ebuild.sh, line   48:  Called src_compile
>>  *             environment, line 2238:  Called die
>>  * The specific snippet of code:
>>  *       make -f Makefile.Generic || die "Compilation failed."
>>  *  The die message:
>>  *   Compilation failed.
>>
>> I also attached the build log and the  ebuild environment file.
>>
>> Any suggestions on how to solve this problem? Kind of urgent cause I
>> need to finish my job-application...
>
> I had the same problem. The fix in bug
> http://bugs.gentoo.org/show_bug.cgi?id=264510 solved it for me.

Since I am rather new to gentoo, I am not sure how to apply the
patches together with emerge. Could you give me a short info on that
or point me to the corresponding documentation/howto?

Thanks

--
Regards,
 Marco

Re: [gentoo-user] [ot] no more inodes

[Neil Bothwick]

On Fri, 29 May 2009 19:39:09 -0600, Maxim Wexler wrote:

> > I found the best way to deal with the Eee 900's two drives was to
> > create a small root partition (I used 200M) and swap on sda. Then
> > make the rest of sda and all of sdb into an LVM volume group. I still
> > use ext3 for /, but it contains so little that inodes are not an
> > issue. You definitely want to get /usr/portage, $PORTAGE_TMPDIR and
> > $DISTDIR off the root partition.  
> 
> Just got back from Circuit City or whatever it's called with a 16G SD
> card and I'm steeling myself for the big task ahead. Just what do you
> have under root? How did you format the rest?

My SD card is not part of the volume group. The Eee PC 900 has two SSDs
internally, one at 4GB and one at 16GB (for the Linux version). The root
partiton only contains what needs to be there: /boot, /etc, /bin, /lib
and /sbin. Everything else (/usr, /var, /home, /opt) is on the VG.
$PORTAGE_TMPDIR and $DISTDIR and on a network mount.


-- 
Neil Bothwick

Sir! Romulan warbird decloaki»®õ÷üÁ NO CARRIER


signature.asc
Description: PGP signature

Re: [gentoo-user] Optimizations for SSD netbook

[Florian Philipp]

Grant schrieb:
> My girlfriend is at her wit's end with her SSD netbook and is now
> hogging my laptop.  Her netbook has 1GB RAM that could be upgraded to
> 1.5GB, but I've read that it's a pain.  It already runs xfce4, and
> I've just made these optimizations based on past discussions:
> 
> 1. CFLAGS="-march=prescott -0s -pipe -fomit-frame-pointer -ssse3"
> 2. added elevator=noop as a boot parameter

I remember that I've given this second advice. Since then I've read in
the German computer journal c't [1] that CFQ has a detection for SSDs
since 2.6.28 and now is the best choice for these devices.

> 3. disabled DRI to save 32MB RAM
> 4. removed the swap partition from /etc/fstab
> 
> Am I missing anything significant?  I've read that it's good to set up
> /tmp in RAM.  How can I do that? In /etc/fstab I have:
> 
> shm /dev/shm tmpfs nodev,nosuid,noexec
> 
> Is that related?
> 

Yup, the entry should read:
tmp /tmp tmpfs default 0 0

The first entry is just a name. You can name it in every way you want it.

For further tweaks: Do you use Firefox? I've read that it uses fsync()
when writing to its sqlite backend. This is a really good thing because
it reduces the risk of loosing data but you might (and can) disable this
to increase performance and reduce wear. I don't have a link at hand but
it shouldn't be to hard to find.

[1]c't 11/2009 page 101

Re: [gentoo-user] update-eix freezes on pro-audio overlay

[Jean-Baptiste Mestelan]

2009/5/30 Andrew Gaydenko :
> Thanks for the suggestion. Have tried just now. Unfortunately, didn't help.

Well, embarrassingly enough, I have just tried syncing this overlay,
and get stuck at 86% too !
So this would mean the overlay SVN has a problem, server-side, I suppose.

Tweaks for SSDs [Was: [gentoo-user] [ot] no more inodes]

[Florian Philipp]

Maxim Wexler schrieb:
> On 5/28/09, Volker Armin Hemmann  wrote:
>> On Donnerstag 28 Mai 2009, Florian Philipp wrote:
>>> Maxim Wexler schrieb:
 Hi group,

 For a netbook 4G SSD. Attempting to install mozilla-firefox. jdk
 fails: No space left on device.

 df -i reveals no more inodes. I reboot thinking this will help. Wrong.
 Lots of 'No space left on device messages'  with reference to
 /var/lib/iinit.d/* in the boot console. And this gem: '*ERROR: local
 is already starting'. And: '*ERROR: netmount is already starting'.

[...]

 I know 4G is pretty small by today's standards but apart from xorg and
 firefox everything else on this unit is command-line type utilities
 and such. That can't account for 4G already.

 Maxim
>>> That you run out of inodes doesn't mean that you run out of physical (or
>>> logical) space on your disk. It just means that you run out of what you
>>> could call file descriptors.
>>>
>>> There is exactly one inode per file which stores meta information about
>>> this file. Ext2-4 have a fixed amount of inodes set when you format the
>>> partition. Reiserfs and JFS create them on the fly and therefore don't
>>> have problems with running out of inodes or wasting space on unused ones.
>>>
>>> Most likely you have a bunch of very small files on our disk, for
>>> example the portage tree. These don't consume much space but a lot of
>>> inodes.
>>>
>>> My advice: Save everything to another disk and then reformat the
>>> partition with a higher amount of inodes. If you use ext2, format it with
>>>
>>> mke2fs -N 732960 /dev/sda2
>>>
>>> This will create a file system with three times as many indoes as you
>>> had before.
>>>
>>> Hope this helps.
>> or don't use extX.
>>
>>
> Ok, thanks everybody, getting ready to dive in and fix this thing. Two
> more questions please:
>
[...]
>
> What's the best fs for a 4G SSD? I picked ext3 because of another eee
> forum post.
>
> Maxim
> 

I just want to point to three blog posts from Theodore Ts'o:

Partioning scheme and formatting tricks for optimal performance:
http://thunk.org/tytso/blog/2009/02/20/aligning-filesystems-to-an-ssds-erase-block-size/

Talk about some general issues (ATA TRIM, mostly):
http://thunk.org/tytso/blog/2009/02/22/should-filesystems-be-optimized-for-ssds/

Making an argument for using journalling filesystems:
http://thunk.org/tytso/blog/2009/03/01/ssds-journaling-and-noatimerelatime/

Of course, T'so talks about an Intel X25-M which is a completely
different beast from those cheap SSDs you find in netbooks.

Delaying commits with ext4 and/or laptop-mode will reduce the wear-down
of your SSD but it might as well freeze your system when the actual
commit takes place because these things tend to have a terribly low
write performance.

In the end it will be a matter of playing with parameters.

[gentoo-user] Re: Sync'ed my ~x86 system yesterday and now resolver stopped working

[Christer Ekholm]

Timur Aydin  writes:

> Hi,
>
> I have synced my ~x86 system yesterday and after it completed, the
> resolver doesn't work for some programs anymore. For example, ping
>  says "unknown host name". It doesn't even contact the dns
> server, which is running on the same host. But dig  works
> fine. Also, using the IP address directly, I can access the internet.
>
> I am suspecting that the new glibc 2.10 is causing this. Anybody else
> having this issue?

I experienced similar problem when I upgraded to glibc-2.10, but the
only thing I had to do was change from 'server' to 'nameserver' in
/etc/resolv.conf (as told by my 'man resolv.conf'). I don't know why I
had only 'server' before, perhaps that was allowed with earlier glibc.

-- 
 Christer

Re: [gentoo-user] update-eix freezes on pro-audio overlay

[Andrew Gaydenko]

On Saturday 30 May 2009 13:24:53 Jean-Baptiste Mestelan wrote:
> 2009/5/30 Andrew Gaydenko 
>
> > ... without CPU eating. Freezing point is shown below. I have observed
> > zynaddsubfx-related message earlier plenty of times but without any
> > freezing. And '86%' (freezing point) is far after zynaddsubfx ebuild
> > parsing (at 54%).
> >
> > Where to dig in?
> > Something python-related?
> > Temporary workaround?
>
> Just asking : have you tried deleting, then re-creating the overlay ?
>
> layman -d pro-audio && layman -a  pro-audio

Thanks for the suggestion. Have tried just now. Unfortunately, didn't help.

Re: [gentoo-user] update-eix freezes on pro-audio overlay

[Jean-Baptiste Mestelan]

2009/5/30 Andrew Gaydenko 
> ... without CPU eating. Freezing point is shown below. I have observed
> zynaddsubfx-related message earlier plenty of times but without any freezing.
> And '86%' (freezing point) is far after zynaddsubfx ebuild parsing (at 54%).
>
> Where to dig in?
> Something python-related?
> Temporary workaround?

Just asking : have you tried deleting, then re-creating the overlay ?

layman -d pro-audio && layman -a  pro-audio

[gentoo-user] update-eix freezes on pro-audio overlay

[Andrew Gaydenko]

... without CPU eating. Freezing point is shown below. I have observed 
zynaddsubfx-related message earlier plenty of times but without any freezing. 
And '86%' (freezing point) is far after zynaddsubfx ebuild parsing (at 54%).

Where to dig in?
Something python-related?
Temporary workaround?

//
update-eix
Reading Portage settings ..
Building database (/var/cache/eix) ..
[0] "gentoo" /usr/portage/ (cache: metadata-flat)
 Reading 100%
[1] "proaudio" /usr/local/portage/layman/pro-audio (cache: parse|ebuild*)
 Reading  54% *
 * ERROR: media-sound/zynaddsubfx-2.2.1-r5 failed.
 * Call stack:
 * ebuild.sh, line 1879:  Called _source_ebuild
 * ebuild.sh, line 1818:  Called source 
'/usr/local/portage/layman/pro-audio/media-
sound/zynaddsubfx/zynaddsubfx-2.2.1-r5.ebuild'
 *   zynaddsubfx-2.2.1-r5.ebuild, line5:  Called inherit 'eutils' 'zyn2'
 * ebuild.sh, line 1272:  Called die
 * The specific snippet of code:
 *  declare -F "${ECLASS}_$x" >/dev/null || \
 *  die "EXPORT_FUNCTIONS: ${ECLASS}_$x is 
not defined"
 *  The die message:
 *   EXPORT_FUNCTIONS: zyn2_src_compile is not defined
 *
 * If you need support, post the topmost build error, and the call stack if 
relevant.
 * This ebuild is from an overlay: '/usr/local/portage/layman/pro-audio/'
 *

Ebuild failed with status 1
 Reading  54%
Could not properly execute /usr/local/portage/layman/pro-audio/media-
sound/zynaddsubfx/zynaddsubfx-2.2.1-r5.ebuild
 Reading  86%

Re: [gentoo-user] pdftk emerge error

[Dan Johansson]

On Saturday 30 May 2009, Marco wrote:
> Hi all,
>
> I am trying to emerge pdftk on my amd64 machine. After re-emerging gcc
> with gcj, compilation fails with:
>
> * ERROR: app-text/pdftk-1.12 failed.
>  * Call stack:
>  *   ebuild.sh, line   48:  Called src_compile
>  * environment, line 2238:  Called die
>  * The specific snippet of code:
>  *   make -f Makefile.Generic || die "Compilation failed."
>  *  The die message:
>  *   Compilation failed.
>
> I also attached the build log and the  ebuild environment file.
>
> Any suggestions on how to solve this problem? Kind of urgent cause I
> need to finish my job-application...

I had the same problem. The fix in bug 
http://bugs.gentoo.org/show_bug.cgi?id=264510 solved it for me.

Regards,
-- 
Dan Johansson, 
***
This message is printed on 100% recycled electrons!
***


smime.p7s
Description: S/MIME cryptographic signature